NAS and arrays · bench log · ADR-2025-3653
Six Disks Out of the Box, and a Volume Reading RAW.
The NAS returned after restarting but the volume did not. There were six bays, 6 x 2TB drives running RAID 5
, and where the share used to reside there was a partition listed as 'RAW' format
. He had correctly identified it himself prior to calling: the file system has become corrupted after being turned on
. What he requested was either repairing the array or, failing that, removing the data from the array, and what he needed to know before doing anything was whether or not to allow the box to repair itself.
Same story on yours? Tell us what it does — the look is free.
0203 0868633
What failed, and why.
"RAW" is not a description of the disks. "RAW" is what an operating system describes a partition as when it sees a partition and has no idea how to interpret the structure within it. So the operating system reports that space as RAW, unformatted space and then offers to format that space, which is the most expensive button on the screen. Below that RAW structure, on all six disks, all the data is still exactly where it was written. The clock for a job like this isn't ours; it is the box's. If a NAS boots and finds its volume unreadable, it will often try to help by performing some kind of consistency check, resync or rebuild onto a spare disk, and every single one of those writes to each member while the structure above them is broken. The array does not get better during this process. In fact, it becomes harder to reconstruct. So the first thing asked of anybody who is in this position is to remove power from the unit and keep the disks in their original bay order: if possible take them out of their bays, each labelled with its bay number, or otherwise leave them in the NAS.
What the bench did with it.
The order of the job →| Equipment used | Its work on this job | The reason for it |
|---|---|---|
| Atola TaskForce 2 | All six members copied side by side, so none waited for another | Copies every member of a set at once, so none sit waiting for another |
| PC-3000 Portable PRO | Checked condition and firmware state of each of the six prior to copying being initiated | Manufacturer-level access to an individual disk, on a unit located beside the imagers |
| UFS Explorer RAID Recovery | Confirmed stripe order and parity rotation from the copies, then repaired the file system above them | Recovers the stripe order, then reverse engineers the volume manager the NAS layered above it |
What the bench did with it.
Copy all six before deciding anything
There was no analysis done on the physical disks themselves. All six disks were placed in the imager at the same time, each read straight through, and the decisions made for the rest of the process were based solely on the copies created. You can put a six-member set through one channel or you can image them simultaneously. If you do them sequentially (one at a time), a job of this size will take most of a week; if you do them simultaneously (all at once), you will have the entire set on the bench within a single working day.
Prove each parameter rather than assuming any of them
You are looking at an enormous number of possibilities when trying to determine how many ways you can arrange six disks. Only one arrangement is correct. Each of these parameters (bay order, stripe width and parity rotation direction) was determined from the copies before being tested against actual directory structure layouts already present on the volume. A candidate which generates readable folders has been validated, but one which simply duplicates what a particular manufacturer's usual practice would dictate does not count. This assembly was performed entirely within the software environment, therefore no bytes were ever written to any of the disks during the process.
Repair the file system that sits above the array
After determining the arrangement of the original disk set, the volume containing the damaged file system was mounted on the reconstructed assembly of the disk set. The damage was found precisely where he had said it would be: in the areas that describe the file system itself, rather than in the data locations pointed to by those descriptions. We were then able to repair the damaged areas on our copies. Next, we traversed the file system starting from the root level downward through the hierarchy, examining it folder by folder and opening files as opposed to simply counting them. This is because even though a file may contain the proper amount of data, its contents may still be incorrect.
What went home.
The volume is now readable again, and it was transferred to new media. All six original disks were returned unchanged and still in their original bay order. Not a byte was ever written to them, and the NAS was never requested to rebuild anything. If it had been allowed to do so, this entry would have been a longer one with a worse outcome.
Begin here if yours is doing the same thing.
Other RAID & NAS work on the log.
Yours doing something similar? Power it down first.
Published prices are fixed. Memory cards and USB sticks are £250 + VAT. Any stand-alone drive is £300 + VAT. Drive configurations containing multiple disks (RAID, NAS or server) start at £500 + VAT. First assessments are always complimentary. Thereafter, data recovery charges are also fixed in advance and in writing, so you'll always know how much you owe before paying any money. Shut it down, get a quote first (by phone or with the form), then send it in or drop it off at our Guildford intake laboratory, Monday to Friday, 9am to 5:30pm (no appointment needed). On most jobs you only pay if we recover your data.